The global burden of trichiasis in 2016

Abstract

Background

Trichiasis is present when one or more eyelashes touches the eye. Uncorrected, it can cause blindness. Accurate estimates of numbers affected, and their geographical distribution, help guide resource allocation.

Methods

We obtained district-level trichiasis prevalence estimates in adults for 44 endemic and previously-endemic countries. We used (1) the most recent data for a district, if more than one estimate was available; (2) age- and sex-standardized corrections of historic estimates, where raw data were available; (3) historic estimates adjusted using a mean adjustment factor for districts where raw data were unavailable; and (4) expert assessment of available data for districts for which no prevalence estimates were available.

Findings

Internally age- and sex-standardized data represented 1,355 districts and contributed 662 thousand cases (95% confidence interval [CI] 324 thousand–1.1 million) to the global total. Age- and sex-standardized district-level prevalence estimates differed from raw estimates by a mean factor of 0.45 (range 0.03–2.28). Previously non- stratified estimates for 398 districts, adjusted by ×0.45, contributed a further 411 thousand cases (95% CI 283–557 thousand). Eight countries retained previous estimates, contributing 848 thousand cases (95% CI 225 thousand-1.7 million). New expert assessments in 14 countries contributed 862 thousand cases (95% CI 228 thousand–1.7 million). The global trichiasis burden in 2016 was 2.8 million cases (95% CI 1.1–5.2 million).

Interpretation

The 2016 estimate is lower than previous estimates, probably due to more and better data; scale-up of trichiasis management services; and reductions in incidence due to lower active trachoma prevalence.

About the Authors:

Rebecca M. Flueckiger

Roles Conceptualization, Data curation, Formal analysis, Methodology, Project administration, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliation: Faculty of Infectious and Tropical Diseases, London School of Hygiene & Tropical Medicine, London, United Kingdom

ORCID logo http://orcid.org/0000-0003-4294-2816

Paul Courtright

Roles Conceptualization, Methodology, Writing – review & editing

Affiliation: Kilimanjaro Centre for Community Ophthalmology, Division of Ophthalmology, University of Cape Town, Cape Town, South Africa

ORCID logo http://orcid.org/0000-0002-0997-347X

Mariamo Abdala

Roles Data curation, Writing – review & editing

Affiliation: Ministerio da Saude, National Ophthalmology Program, Maputo, Mozambique

Amza Abdou

Roles Data curation, Writing – review & editing

Affiliation: Ministère de la Santé et de la population, Niamey, Niger

Zaid Abdulnafea

Roles Validation

Affiliation: Ministry of Health, Baghdad, Iraq

Tawfik K. Al-Khatib

Roles Data curation, Writing – review & editing

Affiliation: National Eye Health Programme, Ministry of Public Health and Population, Sana’a, Yemen

Khaled Amer

Roles Data curation, Writing – review & editing

Affiliation: Ministry of Health, Cairo, Egypt

Olga Nelson Amiel

Roles Data curation, Writing – review & editing

Affiliation: Ministerio da Saude, Maputo, Mozambique

Sossinou Awoussi

Roles Data curation, Writing – review & editing

Affiliation: Ministère de la Santé, Lomé, Togo

Ana Bakhtiari

Roles Data curation, Validation, Writing – review & editing

Affiliation: The Task Force for Global Health, Atlanta, GA, USA

Wilfried Batcho

Roles Data curation, Writing – review & editing

Affiliation: Ministère de la Santé, Cotonou, Benin

Assumpta Lucienne Bella

Roles Data curation, Writing – review & editing

Affiliation: Ministère de la Santé, Yaoundé, Cameroun

Kamal Hashim Bennawi

Roles Data curation, Writing – review & editing

Affiliation: Prevention of Blindness Program, Khartoum, Sudan

Simon J. Brooker

Roles Conceptualization, Methodology, Writing – review & editing

Affiliations Faculty of Infectious and Tropical Diseases, London School of Hygiene & Tropical Medicine, London, United Kingdom, Global Health, Neglected Tropical Diseases, Bill & Melinda Gates Foundation, Seattle, USA

Brian K. Chu

Roles Methodology, Writing – review & editing

Affiliation: The Task Force for Global Health, Atlanta, GA, USA

Michael Dejene

Roles Methodology, Writing – review & editing

Affiliation: Michael Dejene Public Health Consultancy Services, Addis Ababa, Ethiopia

Djore Dezoumbe

Roles Validation

Affiliation: Ministère de la Santé, N'Djamena, Chad

Balgesa Elkheir Elshafie

Roles Data curation, Writing – review & editing

Affiliation: Trachoma Control Program, National Ministry of Health, Khartoum, Sudan

Aba Ange Elvis

Roles Data curation, Writing – review & editing

Affiliation: Programme National de la Santé Oculaire et de la lutte contre l'Onchocercose, Abidjan, Côte d'Ivoire

Djouma Nembot Fabrice

Roles Data curation, Writing – review & editing

Affiliation: Ministère de la Santé, Yaoundé, Cameroun

Fatma Juma Omar

Roles Data curation, Writing – review & editing

Affiliation: Ministry of Health, Zanzibar, United Republic of Tanzania

Missamou François

Roles Data curation, Writing – review & editing

Affiliation: Bureau des Maladies Oculaires, Ministère de la Santé, Kinshasa, Democratic Republic of the Congo

Drabo François

Roles Data curation, Writing – review & editing

Affiliation: Programme national de lutte contre les maladies tropicales négligées (PNMTN), Ouagadougou, Burkina Faso

Jambi Garap

Roles Data curation, Writing – review & editing

Affiliation: National Department of Health, Port Moresby, Papua New Guinea

Michael Gichangi

Roles Data curation, Writing – review & editing

Affiliation: Ministry of Health, Nairobi, Kenya

André Goepogui

Roles Data curation, Writing – review & editing

Affiliation: Programme National de Lutte Contre l’Onchocercose et la Cécité et les Maladies Tropicales Négligées, Conakry, Guinea

Jaouad Hammou

Roles Data curation, Writing – review & editing

Affiliation: Ministry of Health, Rabat, Morocco

Boubacar Kadri

Roles Data curation, Writing – review & editing

Affiliation: Ministère de la Santé et de la population, Niamey, Niger

George Kabona

Roles Data curation, Writing – review & editing

Affiliation: Ministry of Health, Dar es Salaam, United Republic of Tanzania

Martin Kabore

Roles Data curation, Writing – review & editing

Affiliation: l'unité d'élimination du trachome, PNMTN, Ouagadougou, Burkina Faso

Khumbo Kalua

Roles Data curation, Writing – review & editing

Affiliation: Department of Ophthalmology, University of Malawi, College of Medicine, Blantyre, Malawi

Mathias Kamugisha

Roles Data curation, Writing – review & editing

Affiliation: National Institute for Medical Research, Dar es Salaam, United Republic of Tanzania

Biruck Kebede

Roles Data curation, Writing – review & editing

Affiliation: Federal Ministry of Health, Addis Ababa, Ethiopia

Kaba Keita

Roles Data curation, Writing – review & editing

Affiliation: Programme National de Lutte Contre l’Onchocercose et la Cécité et les Maladies Tropicales Négligées, Conakry, Guinea

Asad Aslam Khan

Roles Data curation, Writing – review & editing

Affiliation: Ministry of Health, Islamabad, Pakistan

Genet Kiflu

Roles Data curation, Writing – review & editing

Affiliation: Federal Ministry of Health, Addis Ababa, Ethiopia

Makoy Yibi

Roles Data curation, Writing – review & editing

Affiliation: Ministry of Health, South Sudan

Garae Mackline

Roles Data curation, Writing – review & editing

Affiliation: Ministry of Health, Port Vila, Vanuatu

Colin Macleod

Roles Data curation, Methodology

Affiliation: Faculty of Infectious and Tropical Diseases, London School of Hygiene & Tropical Medicine, London, United Kingdom

Portia Manangazira

Roles Data curation, Writing – review & editing

Affiliation: Department of Epidemiology and Disease Control, Ministry of Health & Child Welfare, Harare, Zimbabwe

Michael P. Masika

Roles Data curation, Writing – review & editing

Affiliation: Ministry of Health, Lilongwe, Malawi

Marilia Massangaie

Roles Data curation, Writing – review & editing

Affiliation: Ministerio da Saude, Department of Neglected Tropical Diseases, Maputo, Mozambique

Takafira Mduluza

Roles Data curation, Writing – review & editing

Affiliation: Department of Biochemistry, University of Zimbabwe, Harare, Zimbabwe

Nabicassa Meno

Roles Data curation, Writing – review & editing

Affiliation: Ministère de la Santé, Bissau, Guinea-Bissau

Nicholas Midzi

Roles Data curation, Writing – review & editing

Affiliation: Department of Medical Microbiology, College of Health Sciences, University of Zimbabwe, Harare, Zimbabwe

Abdallahi Ould Minnih

Roles Data curation, Writing – review & editing

Affiliation: Minister de la Sante, Nouakchott, Mauritania

Sailesh Mishra

Roles Data curation, Writing – review & editing

Affiliation: Nepal Netra Jyoti Sangh, Kathmandu, Nepal

Caleb Mpyet

Roles Data curation, Writing – review & editing

Affiliation: Sightsavers Nigeria, Kaduna, Nigeria & Department of Ophthalmology, Jos University, Jos, Nigeria

Nicholas Muraguri

Roles Data curation, Writing – review & editing

Affiliation: Ministry of Health, Nairobi, Kenya

Upendo Mwingira

Roles Data curation, Writing – review & editing

Affiliation: Ministry of Health, Dar es Salaam, United Republic of Tanzania

Beido Nassirou

Roles Data curation, Writing – review & editing

Affiliation: Ministère de la Santé et de la population, Niamey, Niger

Jean Ndjemba

Roles Data curation, Writing – review & editing

Affiliation: Bureau des Maladies Oculaires, Ministère de la Santé, Kinshasa, Democratic Republic of the Congo

Cece Nieba

Roles Data curation, Writing – review & editing

Affiliation: Programme National de Lutte Contre l’Onchocercose et la Cécité et les Maladies Tropicales Négligées, Conakry, Guinea

Jeremiah Ngondi

Roles Data curation, Writing – review & editing

Affiliation: RTI International, Dar es Salaam, United Republic of Tanzania

ORCID logo http://orcid.org/0000-0003-4357-4816

Nicholas Olobio

Roles Data curation, Writing – review & editing

Affiliation: Nigeria Federal Ministry of Health, Abuja, Nigeria

Alex Pavluck

Roles Methodology, Writing – review & editing

Affiliation: RTI International, Washington DC, USA

Isaac Phiri

Roles Data curation, Writing – review & editing

Affiliation: Department of Epidemiology and Disease Control, Ministry of Health & Child Welfare, Harare, Zimbabwe

Rachel Pullan

Roles Methodology, Writing – review & editing

Affiliation: Faculty of Infectious and Tropical Diseases, London School of Hygiene & Tropical Medicine, London, United Kingdom

Babar Qureshi

Roles Data curation, Writing – review & editing

Affiliation: Christian Blind Mission, Bensheim, Germany

Boubacar Sarr

Roles Data curation, Writing – review & editing

Affiliation: Ministère de la santé et de l’Action Sociale, Dakar, Senegal

Do Seiha

Roles Data curation, Writing – review & editing

Affiliation: Prevention of Blindness Programme, Ministry of Health, Phnom Penh, Cambodia

Gloria Marina Serrano Chávez

Roles Data curation, Writing – review & editing

Affiliation: Ministry of Health, Guatemala City, Guatemala

Shekhar Sharma

Roles Data curation, Writing – review & editing

Affiliation: Ministry of Health, Kathmandu, Nepal

Siphetthavong Sisaleumsak

Roles Data curation, Writing – review & editing

Affiliation: National Ophthalmology Centre, Vientiane, Lao People’s Democratic Republic

Khamphoua Southisombath

Roles Conceptualization, Data curation, Writing – review & editing

Affiliation: National Program for the Prevention of Blindness, Ministry of Health, Vientiane, Lao People’s Democratic Republic

Gretchen Stevens

Roles Methodology

Affiliation: Department of Information, Evidence and Research, World Health Organization, Geneva, Switzerland

Andeberhan Tesfazion Woldendrias

Roles Data curation, Writing – review & editing

Affiliation: Department of Public Health, Ministry of Health, Asmara, Eritrea

Lamine Traoré

Roles Data curation, Writing – review & editing

Affiliation: Ministère de la Santé, Bamako, Mali

Patrick Turyaguma

Roles Data curation, Writing – review & editing

Affiliation: Trachoma Program, Ministry of Health, Kampala, Uganda

Rebecca Willis

Roles Methodology, Writing – review & editing

Affiliation: The Task Force for Global Health, Atlanta, GA, USA

Georges Yaya

Roles Data curation, Writing – review & editing

Affiliation: Ministère de la Santé Publique, Bangui, Central African Republic

Souleymane Yeo

Roles Data curation, Writing – review & editing

Affiliation: Programme National de la Santé Oculaire et de la lutte contre l'Onchocercose, Abidjan, Côte d'Ivoire

Francisco Zambroni

Roles Conceptualization, Data curation, Writing – review & editing

Affiliation: Ministry of Health, Guatemala City, Guatemala

Jialiang Zhao

Roles Data curation, Writing – review & editing

Affiliation: Department of Ophthalmology, Peking Union Medical Colllege Hospital, Chinese Academy of Medical Sciences, Beijing, China

Anthony W. Solomon

Roles Conceptualization, Methodology, Resources, Writing – review & editing

Affiliation: Department of Control of Neglected Tropical Diseases, World Health Organization, Geneva, Switzerland

Introduction

Trachoma, a neglected tropical disease, is endemic or has recently been endemic in more than 50 countries[1]. It affects the most impoverished people of the world. Improved living standards are credited for the disappearance of trachoma from Europe and North America, but in many less developed countries, trachoma is still a public health problem, and contributes to the continued suffering and deepening of poverty of millions of people.

Chlamydia trachomatis is the causative organism. Repeated ocular chlamydial infection results in chronic inflammation, characterised by sub-epithelial follicles in the tarsal conjunctiva, which may be sufficiently large and numerous to meet the definition of trachomatous inflammation—follicular (TF), a key sign for assessing trachoma prevalence[2]. Over time, with repeated reinfection, scarring may develop; this scarring can eventually cause the eyelid to turn inwards in some people, resulting in eyelashes touching the globe. This is called trachomatous trichiasis (TT) and is very painful[3]. As an individual with trichiasis blinks, the eyelashes abrade the cornea, which can lead to corneal opacity and blindness[4].

The World Health Organization (WHO) endorses population based prevalence surveys (PBPSs) to guide decisions regarding implementation of interventions for trachoma elimination[5]. The typical survey design involves two-stage cluster sampling, which uses probability-proportional-to-size methodologies to select 20–30 clusters. The outputs are estimates of the prevalence of TF in children aged 1–9 years, and the prevalence of TT in adults aged 15 years and older[6].

Through high quality surgery, which involves altering the position of the eyelid margin,[7] it is possible to reduce the number of people with TT. Accurate estimates of the number of persons with TT (the TT backlog) and their geographical distribution are needed in order to effectively align resources for surgery and other necessary services.

In 2009, Mariotti et al estimated the global TT backlog to be 8.2 million people[8]. This estimate was derived by summarizing a combination of published and unpublished information. First, a literature review was performed to identify published prevalence data. Where published data were not available, unpublished data were compiled from the Eleventh (2007) Meeting of the WHO Alliance for the Global Elimination of Trachoma by 2020[9]. Where information was still missing, unpublished reports were provided by health ministries. Finally, if none of these resources were available, data were extrapolated from a proxy country believed to have common epidemiological conditions and demographic structure.

There are many uncertainties inherent in the 2009 estimate[8]. First, where PBPS data were available, the results were not systematically stratified by age and sex. This is problematic because women and the elderly are more likely than men and younger adults, respectively, to (1) be at home at the time that a house-to-house survey team calls, and (2) have TT. Second, in countries for which data were available, survey coverage was generally far from complete. Though not an invariable rule, surveys to estimate the prevalence of neglected diseases have a tendency to be done first in areas of high prevalence; for the 2009 estimate, if any data were available for a particular country, the TT prevalence figure from those data was applied across the yet-to-be-mapped suspected-endemic population. Third, the use of proxy countries is extremely subjective.

In 2012, WHO collected and collated provisional 2011 country reports, and published an updated figure for the global TT backlog. The total given was 7.3 million people, but the methodologies used in each country to generate national backlog figures were not described[10]; they are likely to have been highly heterogeneous.

From December 2012 to January 2016, the Global Trachoma Mapping Project (GTMP) sought to systematically complete the global trachoma map using standardized techniques for both collecting and analysing survey data[11]. Eligibility[12] and protocols [11]for mapping were determined by expert working groups. GTMP measured trachoma prevalence using gold standard PBPSs conducted at district level. People of all ages living in selected households of 1,546 districts across 29 countries were examined for trachoma, resulting in the examination of 2.6 million people[13]. GTMP analyses included stratifying trichiasis prevalence estimates against national population pyramids in an attempt to partially account for demographic differences between those examined and the national averages.

As a result of the GTMP, there are now high quality PBPS data available for most suspected-endemic areas that were previously unsurveyed. The availability of these data has catalysed the current attempt to generate a new estimate of the global trichiasis backlog.

Methods

We updated previous global estimates using the best available data, according to the following hierarchy, with all areas worldwide that have trichiasis prevalence estimates eligible for inclusion. First, where GTMP data[11] were available, we used the age- and sex- stratified survey-level trichiasis prevalence estimates for adults. In August 2014, GTMP added examination for the presence or absence of trachomatous scarring (TS)[2] for all eyes determined to have trichiasis; prior to this, and for all non-GTMP data, trichiasis prevalence estimates included trichiasis with and without TS. All analyses of trichiasis were undertaken among adults aged 15 years and older only.

Second, where PBPSs had been done without the support of the GTMP, we requested de-identified individual-level raw survey data from national health ministries, or appropriate government-designated agencies. Where those data were provided by 1 March 2016, we applied the same age- and sex- stratification as was used in the GTMP[11].

Third, where PBPSs had been done but raw data were not available, survey-level prevalence estimates (whether stratified by age and sex or not) were obtained from country programs. When data had not been stratified, estimates were multiplied by the mean adjustment factor for districts in which age- and sex- stratification was possible.

Fourth, if prevalence estimates were not available, we reviewed previous estimates through desk reviews and communications with country programs. If there was adequate evidence to revise the estimates, new estimates were used; otherwise, the 2009 estimates[8] were retained.

The population data used in these analyses were derived from the UN population division (UNdata)[14] and www.worldpop.org[15]. As trachoma is a disease primarily affecting rural populations[4], we used rural population estimates for this purpose. Rural population pyramids were obtained from UNdata. Microsoft Excel (2007) was used to organize the datasets into 5-year age bands stratified by sex for each country. The percentage of the population within each stratum was estimated from the bands. The district level populations were derived from www.worldpop.org raster files[15] using the zonal statistics tool in ArcGIS 10.3[16]. The raster files used here are a matrix of geolocated pixels where each pixel contains population values. The sum values of the pixels falling within each district boundary was taken as the district value. A sensitivity analysis was also performed, comparing the www.worldpop.org population estimates to district population estimates provided by national trachoma elimination programs. The mean ratio between the national program estimates and www.worldpop.org was 0·97. Because of this close correlation, and for the purposes of standardizing our methods, the www.worldpop.org data were used throughout this analysis.

For population-based prevalence surveys done without GTMP support but for which raw data were available, the statistical software package R[17] was used to perform age- and sex- post-stratification. First, the crude prevalence was calculated for each cluster. Second, the prevalence was stratified for each cluster by weighting the proportion of each sex-specific five-year age band observed to have trichiasis by the proportion of the adults aged 15 years and older expected to have that age and sex in that district, if available, or (if not available), nation-wide. Third, the un-weighted arithmetic mean of the stratified cluster-level trichiasis proportions was taken as the district-level prevalence. Sampling weights were not needed here as the surveys used probability-proportional-to-size sampling and are therefore self-weighted. Once the initial post-stratification was completed, bootstrapping was undertaken on the dataset for each individual district to derive 95% confidence intervals (CIs). For a district surveyed by examining individuals in n clusters, this involved bootstrap resampling (with replacement) of n clusters, over 10,000 replications. The R code is provided in the Appendix.

Where PBPSs had been done but raw data were not available, CIs were constructed around the original estimate by applying an approximation to the binomial distribution , where p is the proportion of positive cases in the sample, s.e. is , and is the standard normal distribution (95% CI = 1.96). Where the sample size was not available, a sample of 1000 was assumed. This method was used at the district-level when data were available and country-level when previous estimates were retained. When the district-level prevalence estimate was multiplied by the mean adjustment factor for districts in which age- and sex- stratification was possible, the upper and lower bounds of the corresponding confidence intervals were adjusted in the same way.

Post-stratified prevalence estimates and the lower and upper 95% CI bounds were multiplied by the 15-years-and-older population for the relevant district (projected to 2016) to provide an estimate of the number of persons with trichiasis.

Health ministries of endemic countries reviewed the output; for three countries, as will be noted in the results section, the estimate was subsequently reduced to account for implementation of trichiasis surgery programs since the time of survey. This changed the category of estimates in the countries involved from estimates based on PBPS data to “expert assessments”.

For expert backlog assessments, confidence intervals were constructed using the relative precisions of all datasets for which bootstrapped CIs could be determined. The range of relative upper bound and lower bound precisions were trimmed to exclude 10% of the greatest and least values. The means of the remainder were applied.

Upper (and lower) CI bounds around district-level estimates, regardless of the method used for calculation, were summed to generate country-level, region-level and global estimates of the upper (and lower) CI bounds.

A sensitivity analysis was performed where communications with country programs estimated a value of zero. This involved selecting proxy countries based on similar geography and hygiene and sanitation situations and calculating the ratio of estimated country backlog over country population. This ratio was then applied to the zero value countries to estimate a backlog. Hygiene and sanitation data were derived from the WHO/UNICEF Joint Monitoring Program for Water Supply, Sanitation and Hygiene (JMP) [18].

The study was approved by the Research Ethics Committee of the London School of Hygiene & Tropical Medicine (11208). The corresponding author had full access to all the data and had final responsibility for the decision to submit for publication.

Results

The 2009 estimate included 57 historically trachoma-endemic countries; this updated estimate reviewed 65 known and suspected trachoma-endemic countries. Descriptive statistics for cluster level data are found in the individual country-level publications. Data from 1,355 districts in 31 countries were age- and sex- stratified and contribute an estimated 662 thousand cases (95% confidence interval (CI) 324 thousand-1.1 million) to the global total. These data were made up of 29,847 clusters, averaging 22 clusters per district. Adjusting prevalence estimates by age and sex reduced raw district-level estimates by a mean factor of 0.45 (range 0.03–2.28; Fig 1). This value was calculated by averaging the ratio of adjusted and unadjusted prevalence estimates across districts that had raw data available. While the district-level estimates reduce by a mean factor of 0.45, the overall country-level backlog reduced by a mean factor of 0.56 (range 0.22–0.94) (Table 1).

[Figure omitted. See PDF.]

Fig 1. Histogram for ratio of adjusted and unadjusted prevalence estimates.

https://doi.org/10.1371/journal.pntd.0007835.g001

[Figure omitted. See PDF.]

Table 1. Estimated national-level trichiasis backlogs, 2016, with comparisons to the corresponding estimates for 2009.

https://doi.org/10.1371/journal.pntd.0007835.t001

Estimates from PBPSs in 398 districts could not be stratified by age and sex because the original datasets were unavailable; the unadjusted backlogs in these districts totalled 781 thousand cases. We adjusted these estimates by multiplying them across the board by 0.45, yielding 411 thousand cases (95% CI 283–557 thousand).

Based on new expert assessment of available data, it was determined that eight countries no longer have a trichiasis backlog. These countries were Angola, Botswana, Burundi, Iran, Iraq, Namibia, Nauru, and Mexico. For six countries—Djibouti, Ethiopia, Libya, Mali, Niger and South Sudan—local expert assessment generated revised, non-zero estimates. In Ethiopia, Niger and Mali, the local expert assessment was undertaken at the request of the health ministry, where, for each district the number of surgeries done since the most recent prevalence survey (as reported annually to WHO) was subtracted from the estimated district-level backlog. (Incorporating surgical delivery in these three countries reduced the global backlog estimate by 124,433 cases: Ethiopia 69,133, Mali 32,083, and Niger 23,217.) For one country, other published estimates[19] were used.

Finally, there were seven countries for which the 2009 estimates were retained. The country-level data derived from previous estimates and expert assessments contributes 61.4% of the overall estimate (Table 2 and Fig 2). Data included in the overall estimate had been collected from 2000–2016 (Fig 3).

[Figure omitted. See PDF.]

Fig 2. Distribution of analysis methodology.

https://doi.org/10.1371/journal.pntd.0007835.g002

[Figure omitted. See PDF.]

Fig 3. Number of districts contributing data to the overall estimate, by year of survey.

https://doi.org/10.1371/journal.pntd.0007835.g003

[Figure omitted. See PDF.]

Table 2. Backlog estimate attributed to each analysis methodology.

https://doi.org/10.1371/journal.pntd.0007835.t002

Through the inclusion of GTMP data, adjusting older datasets by age and sex, and obtaining current local expert assessment of available data, the global backlog estimate reduces from 2009’s 8.2 million[8] (or 2012’s 7.3 million[10]) to 2.8 million (95% CI 1.1–5.2 million) in 2016 (Tables 2 and 3). The updated estimate reduced to zero cases in twelve countries, namely: Angola, Botswana, Burundi, Congo, Fiji, Gambia, Iraq, Islamic Republic of Iran, Morocco, Mexico, Namibia, and Nauru. (Updated survey data demonstrated zero prevalence in Congo, Gambia, Morocco and Fiji, while health ministries in Angola, Botswana, Burundi, Iraq, Islamic Republic of Iran, Mexico, Namibia and Nauru reported no current evidence of trichiasis.) The sensitivity analysis performed for countries where the health ministry reported no current evidence of trichiasis suggests these zero estimates may contribute to a global underestimate by a margin of 13,285, a small difference in the context of the overall estimate, particularly given uncertainty from other sources. The outputs from this analysis are provided in the Appendix.

[Figure omitted. See PDF.]

Table 3. Estimated region-level trichiasis backlogs, 2016, with comparisons to the corresponding estimates for 2009.

https://doi.org/10.1371/journal.pntd.0007835.t003

In our estimates, data collected in surveys set up prior to August 2014 include all trichiasis, irrespective of the presence or absence of trachomatous conjunctival scarring. Data collected in surveys set up after August 2014 contributed data on trichiasis in eyes that also demonstrated trachomatous conjunctival scarring (or had an eyelid that could not be everted, due to presumed dense scar), thereby including only those cases of trichiasis attributable to trachoma. In both scenarios the data represent both “managed” and “unmanaged” trichiasis, irrespective of whether individuals have previously been offered corrective surgery or epilation[45].

Discussion

Trichiasis remains a significant public health problem in many countries, with a global backlog estimated for 2016 at 2.8 million people, 61% of whom live in sub-Saharan Africa. While there were methodological challenges in generating the 2009 estimate, it was the best estimate that could be made with the information then available. The considerable reduction from that estimate to the one generated here is likely to be the result of a combination of factors. First, there are now more and better data derived from rigorous surveys. Second, in many countries, there has been impressive programmatic scale up to manage TT, conducted by complex networks of governments and their partners. Third, there is likely to be an effect on the incidence of TT from the intensive efforts to reduce active trachoma prevalence in many contexts; such efforts have been ramping up in endemic countries since the World Health Assembly’s 1998 commitment to global elimination of trachoma[46]. Teasing out the relative contribution of each of these factors is not possible at the present time, but regardless of cause, the reduction is welcome news for global health.

A better understanding of the backlog and distribution of trichiasis cases is necessary to effectively plan for surgical services and other components of management of individuals with trichiasis. To reduce TT prevalence in each district of each endemic country to <0.2% in adults, which is the defined elimination threshold for TT[47], at least 2.0 million people will need to have their TT appropriately managed. From available data, it is estimated that 56% of people with trichiasis have bilateral trichiasis.

While our calculations lessen the uncertainty around the global backlog estimate, important limitations remain. Expert assessment of available data was used for 14 endemic countries, in eight of which the estimate was zero cases. The sensitivity analysis performed on these eight countries suggests that the zero estimates could contribute to global underestimation by a margin of 13,285. Ethiopia, Mali, and Niger have undergone intensive surgical scale-up in the time since the most recent round of prevalence surveys. Because of this programmatic output, and a lack of consensus around how to counterbalance backlog reduction with new incident cases and post-surgical recurrence (both of which are inescapable, but presently difficult to accurately quantify) these countries provided results based on expert assessment—albeit assessments grounded on 196 (Ethiopia), 55 (Mali) and 49 (Niger) PBPS datasets, and corresponding district-level data on surgical output. Uncertainty is greatest among the eight countries in which previous estimates were retained; these countries accounted for 848 thousand cases of trichiasis (31% of the global estimate) and further investigation is needed here. India, which accounts for 443,000 cases of trichiasis (52% of the total retained estimate) is a particular priority. A second ongoing uncertainty stems from the unavailability of raw data for 398 districts, for which, as a result, age- and sex- stratification was not possible. As noted in our calculations, there is a mean prevalence reduction against the raw prevalence estimate of ×0·45, which was applied to previously-unstratified district-level estimates. Third, our CIs reflect statistical uncertainty. Because statistical uncertainty is uncorrelated across countries, our method of aggregating CIs across administrative boundaries overestimates statistical uncertainty at regional and global levels. However, there are other sources of uncertainty in addition to statistical uncertainty, so it is not possible to say whether our regional and global CIs are biased. Fourth, we have implicitly assumed the prevalence of TT in those aged <15 years to be negligible. This is generally valid, but was a necessary assumption because of the unavailability of data on paediatric trichiasis in many pre-GTMP surveys. Fifth, it is possible that there are some population-based data on trichiasis of which we are unaware, but the partnerships engendered within the WHO Alliance for GET2020 [48]and the existence of the Trachoma Atlas (www.trachomaatlas.org) seem to us to make it unlikely that there are many estimates in this category. Finally, some of the surveys contributing data were conducted more than a decade ago (Fig 2). Although we note above that reductions in active trachoma prevalence are likely to have reduced the incidence of TT over time, we cannot adjust for such reductions within our data. Similarly, we are unable to adjust for the effects of all-cause mortality in patients with TT, population ageing, and other broad demographic changes that have occurred since raw data were collected.

Ongoing lack of trichiasis (and TF) prevalence data for 235 suspected-endemic districts is due to local insecurity, and the global trachoma community stands ready to support national governments to undertake the needed mapping in those populations, when and if security conditions improve to allow safe conduct of fieldwork.

It is recognized[11] that the PBPS methodology described earlier is potentially imprecise in estimating TT against the WHO elimination target of ≤0.2% in adults. Thus, at district level, the uncertainty around the estimates of the TT backlog can be large. This is reflected in our CIs. Work is underway to try to develop more reliable methodologies for estimating the prevalence of trichiasis.

At present, the diagnosis of TT is based on the presence of clinical trichiasis (one or more lashes touching the globe or evidence of recent epilation of in-turned eyelashes) plus residence in a (presumed) trachoma-endemic setting. This definition is by nature somewhat circular and may need review, as it inevitably leads to classification of trichiasis as TT regardless of whether trachomatous conjunctival scarring is present or not. Trichiasis without trachomatous conjunctival scarring may be age-related or due to trauma, distichiasis, epiblepharon or other inflammatory disease[49]; the sight-threatening potential and optimal management strategies for non-trachomatous trichiasis still require further investigation[50].

Assumptions were made when adjusting the data for age and sex. First, we assumed that ages reported in surveys were accurate. However, it is reasonable to hypothesise that during a survey, individuals demonstrate a terminal digit preference when stating their age. Second, this analysis assumed that UNdata population pyramids used were representative of all districts for which survey data were available. Finally, as already noted, we assumed zero trichiasis cases in the population aged 14 years and younger; studies demonstrate a high correlation between trichiasis and increased age[51–54].

These calculations of national and global TT burdens are point-prevalence estimates based on data of differing vintages; they will change as additional surveys, baseline or impact, are undertaken. Other than for Ethiopia, Mali and Niger, the estimates do not take into account the number of surgeries done during the period between the most recent survey and the cut-off date for this analysis. Nevertheless, estimates of current national TT backlog are essential for countries to appropriately allocate resources for surgical campaigns. Future district-level impact surveys in countries with trachoma elimination programs will provide progressively improved evidence for countries to assess their residual TT burden and, ultimately, validate elimination of trichiasis as a public health problem. Baseline surveys of trachoma are still needed in a handful of countries (e.g., Egypt, Somalia and Central African Republic); these will lead to revisions in national estimates. In some settings, there will be a need to undertake trichiasis-only surveys [55] in order to assess progress to elimination. Unfortunately, insecurity may continue to limit surveys in a few, but not all, of the settings in which previous estimates were retained. Where possible, surveys should be undertaken. Lack of primary data will hamper progress to the declaration of global elimination of trachoma as a public health problem. As new primary survey findings become available it will be possible to update global figures. Modelling may also be an important tool for addressing this challenge. Continued efforts by national health ministries, WHO, program partners and researchers to carry out baseline, impact, and pre-surveillance surveys drive revisions to the global elimination of trachomatous trichiasis as a public health problem.

Supporting information

S1 Appendix. R script for calculating TT prevalence normalized by age and sex.

https://doi.org/10.1371/journal.pntd.0007835.s001

(DOCX)

S2 Appendix. TT backlog sensitivity analysis.

https://doi.org/10.1371/journal.pntd.0007835.s002

(DOCX)

S1 Checklist. STROBE checklist.

https://doi.org/10.1371/journal.pntd.0007835.s003

(DOCX)

Citation: Flueckiger RM, Courtright P, Abdala M, Abdou A, Abdulnafea Z, Al-Khatib TK, et al. (2019) The global burden of trichiasis in 2016. PLoS Negl Trop Dis 13(11): e0007835. https://doi.org/10.1371/journal.pntd.0007835

References

1. World Health Organization. WHO Alliance for the Global Elimination of Blinding Trachoma by the year 2020: progress report on elimination of trachoma, 2013. Wkly Epidemiol Rec. 2014;89(39):421–8. pmid:25275153

2. Thylefors B, Dawson CR, Jones BR, West SK, Taylor HR. A simple system for the assessment of trachoma and its complications. Bulletin of the World Health Organization. 1987;65(4):477–83. pmid:3500800; PubMed Central PMCID: PMC2491032.

3. Palmer SL, Winskell K, Patterson AE, Boubacar K, Ibrahim F, Namata I, et al. 'A living death': a qualitative assessment of quality of life among women with trichiasis in rural Niger. International health. 2014;6(4):291–7. pmid:25125577.

4. Mabey DC, Solomon AW, Foster A. Trachoma. Lancet. 2003;362(9379):223–9. pmid:12885486.

5. Solomon AW ZM, Kuper H, et al. Trachoma control: a guide for program managers. Geneva: World Health Organization, 2006.

6. Solomon AW, Peeling RW, Foster A, Mabey DC. Diagnosis and assessment of trachoma. Clinical microbiology reviews. 2004;17(4):982–1011. pmid:15489358; PubMed Central PMCID: PMC523557.

7. Merbs SL, Resnikoff S, Kello AB, Mariotti S, Greene G, West SK. Trichiasis surgery for trachoma. Geneva, Switzerland: World Health Organization, 2015.

8. Mariotti SP, Pascolini D, Rose-Nussbaumer J. Trachoma: global magnitude of a preventable cause of blindness. Br J Ophthalmol. 2009;93(5):563–8. pmid:19098034.

9. WHO Alliance for the Global Elimination of Blinding Trachoma by 2020. Report of the 11th meeting of the WHO Alliance for the Global Elimination of Blinding Trachoma. Geneva, Switzerland: 2007 2–4 April 2007.

10. World Health Organization. Global WHO Alliance for the Elimination of Blinding Trachoma by 2020. Wkly Epidemiol Rec. 2012;87(17):161–8. pmid:22574352

11. Solomon AW, Pavluck AL, Courtright P, Aboe A, Adamu L, Alemayehu W, et al. The Global Trachoma Mapping Project: Methodology of a 34-Country Population-Based Study. Ophthalmic Epidemiol. 2015;22(3):214–25. pmid:26158580.

12. Solomon AW, Willis R, Pavluck AL, Alemayehu W, Bakhtiari A, Bovill S, et al. Quality Assurance and Quality Control in the Global Trachoma Mapping Project. Am J Trop Med Hyg. 2018;99(4):858–63. pmid:30039782; PubMed Central PMCID: PMC6159583.

13. Strachan CE. Independent report: evaluation of Global Trachoma Mapping Project [available at: https://www.gov.uk/government/publications/evaluation-of-global-trachoma-mapping-project, accessed 17 August 2017]. London: Department for International Development; 2017.

14. UNdata: UNSD Demographic Statistics. 0.14.6 ed. UN population division.

15. Worldpop: High resolution age-structured populatin distribution maps [Internet]. GeoData Institute, University of Southampton. [cited 12 January 2016]. Available from: www.worldpop.org.

16. ArcGIS Desktop. 10.3 ed. Redlands, CA: Environmental Systems Research Institute (ESRI); 2014.

17. R Development Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing; 2016.

18. Progress on drinking water, sanitation and hygiene: 2017 update and SDG baseline. Geneva: World Health Organization (WHO) and the United Nations Children’s Fund (UNICEF), 2017 Licence: CC BY-NC-SA 3.0 IGO.

19. The end in sight: 2020 INSight. International Coalition for Trachoma Control, 2011.

20. Bio AA, Boko PM, Dossou YA, et al. Prevalence of trachoma in northern Benin: results from 11 population-based prevalence surveys covering 26 districts. Ophthalmic Epidemiol 2017; 24(4): 265–73. pmid:28441120

21. Kilangalanga J, Ndjemba JM, Uvon PA, et al. Trachoma in the Democratic Republic of the Congo: results of 46 baseline prevalence surveys conducted with the Global Trachoma Mapping Project. Ophthalmic Epidemiol 2018; 25(sup1): 192–200. pmid:28850014

22. Abashawl A, Macleod C, Riang J, Mossisa F, Dejene M, Willis R, et al. Prevalence of Trachoma in Gambella Region, Ethiopia: Results of Three Population-Based Prevalence Surveys Conducted with the Global Trachoma Mapping Project. Ophthalmic Epidemiol. 2016;23(sup1):77–83. pmid:27918222.

23. Adamu Y, Macleod C, Adamu L, Fikru W, Kidu B, Abashawl A, et al. Prevalence of Trachoma in Benishangul Gumuz Region, Ethiopia: Results of Seven Population-Based Surveys from the Global Trachoma Mapping Project. Ophthalmic Epidemiol. 2016;23(sup1):70–6. pmid:27918248.

24. Adera TH, Macleod C, Endriyas M, Dejene M, Willis R, Chu BK, et al. Prevalence of and Risk Factors for Trachoma in Southern Nations, Nationalities, and Peoples' Region, Ethiopia: Results of 40 Population-Based Prevalence Surveys Carried Out with the Global Trachoma Mapping Project. Ophthalmic Epidemiol. 2016;23(sup1):84–93. pmid:27918229.

25. Bero B, Macleod C, Alemayehu W, Gadisa S, Abajobir A, Adamu Y, et al. Prevalence of and Risk Factors for Trachoma in Oromia Regional State of Ethiopia: Results of 79 Population-Based Prevalence Surveys Conducted with the Global Trachoma Mapping Project. Ophthalmic Epidemiol. 2016;23(6):392–405. pmid:27820657.

26. Sherief ST, Macleod C, Gigar G, Godefay H, Abraha A, Dejene M, et al. The Prevalence of Trachoma in Tigray Region, Northern Ethiopia: Results of 11 Population-Based Prevalence Surveys Completed as Part of the Global Trachoma Mapping Project. Ophthalmic Epidemiol. 2016;23(sup1):94–9. pmid:27918232.

27. Kalua K, Chisambi A, Chinyanya D, Kamwendo Z, Masika M, Willis R, et al. Completion of Baseline Trachoma Mapping in Malawi: Results of Eight Population-Based Prevalence Surveys Conducted with the Global Trachoma Mapping Project. Ophthalmic Epidemiol. 2016;23(sup1):32–8. pmid:27726469.

28. Kalua K, Phiri M, Kumwenda I, Masika M, Pavluck AL, Willis R, et al. Baseline Trachoma Mapping in Malawi with the Global Trachoma Mapping Project (GTMP). Ophthalmic Epidemiol. 2015;22(3):176–83. pmid:26158575; PubMed Central PMCID: PMC4673584.

29. Abdala M, Singano CC, Willis R, et al. The epidemiology of trachoma in Mozambique: results of 96 population-based prevalence surveys. Ophthalmic Epidemiol 2017; 25(sup1): 201–10. pmid:28910562

30. Muhammad N, Mpyet C, Adamu MD, William A, Umar MM, Goyol M, et al. Mapping Trachoma in Kaduna State, Nigeria: Results of 23 Local Government Area-Level, Population-Based Prevalence Surveys. Ophthalmic Epidemiol. 2016;23(sup1):46–54. pmid:27918227.

31. Adamu MD, Mpyet C, Muhammad N, Umar MM, Muazu H, Olamiju F, et al. Prevalence of Trachoma in Niger State, North Central Nigeria: Results of 25 Population-Based Prevalence Surveys Carried Out with the Global Trachoma Mapping Project. Ophthalmic Epidemiol. 2016;23(sup1):63–9. pmid:27918223.

32. Mpyet C, Muhammad N, Adamu MD, Muazu H, Umar MM, Abdull M, et al. Prevalence of Trachoma in Bauchi State, Nigeria: Results of 20 Local Government Area-Level Surveys. Ophthalmic Epidemiol. 2016;23(sup1):39–45. pmid:27846369.

33. Mpyet C, Muhammad N, Adamu MD, Muazu H, Umar MM, Goyol M, et al. Prevalence of Trachoma in Katsina State, Nigeria: Results of 34 District-Level Surveys. Ophthalmic Epidemiol. 2016;23(sup1):55–62. pmid:27775463.

34. Mpyet C, Muhammad N, Adamu MD, Muazu H, Umar MM, Alada J, et al. Trachoma Mapping in Gombe State, Nigeria: Results of 11 Local Government Area Surveys. Ophthalmic Epidemiol. 2016;23(6):406–11. pmid:27726459.

35. Mwingira UJ, Kabona G, Kamugisha M, Kirumbi E, Kilembe B, Simon A, et al. Progress of Trachoma Mapping in Mainland Tanzania: Results of Baseline Surveys from 2012 to 2014. Ophthalmic Epidemiol. 2016;23(6):373–80. pmid:27775455; PubMed Central PMCID: PMC5116913.

36. Omar FJ, Kabona G, Abdalla KM, Mohamed SJ, Ali SM, Ame SM, et al. Baseline Trachoma Surveys in Kaskazini A and Micheweni Districts of Zanzibar: Results of Two Population-Based Prevalence Surveys Conducted with the Global Trachoma Mapping Project. Ophthalmic Epidemiol. 2016;23(6):412–7. pmid:27775454; PubMed Central PMCID: PMC5116987.

37. Elshafie BE, Osman KH, Macleod C, Hassan A, Bush S, Dejene M, et al. The Epidemiology of Trachoma in Darfur States and Khartoum State, Sudan: Results of 32 Population-Based Prevalence Surveys. Ophthalmic Epidemiol. 2016;23(6):381–91. pmid:27841721.

38. Meng N, Seiha D, Thorn P, Willis R, Flueckiger RM, Dejene M, et al. Assessment of Trachoma in Cambodia: Trachoma Is Not a Public Health Problem. Ophthalmic Epidemiol. 2016;23(sup1):3–7. pmid:27726472.

39. Macleod CK, Butcher R, Mudaliar U, Natutusau K, Pavluck AL, Willis R, et al. Low Prevalence of Ocular Chlamydia trachomatis Infection and Active Trachoma in the Western Division of Fiji. PLoS Negl Trop Dis. 2016;10(7):e0004798. pmid:27404379; PubMed Central PMCID: PMC4942140.

40. Macleod C, Yalen C, Butcher R, Mudaliar U, Natutusau K, Rainima-Qaniuci M, et al. Eyelash Epilation in the Absence of Trichiasis: Results of a Population-Based Prevalence Survey in the Western Division of Fiji. PLoS Negl Trop Dis. 2017;11(1):e0005277. pmid:28114364; PubMed Central PMCID: PMC5256864.

41. Southisombath K, Sisalermsak S, Chansan P, Akkhavong K, Phommala S, Lewallen S, et al. National Trachoma Assessment in the Lao People's Democratic Republic in 2013–2014. Ophthalmic Epidemiol. 2016;23(sup1):8–14. pmid:27846362.

42. Ko R, Macleod C, Pahau D, Sokana O, Keys D, Burnett A, et al. Population-Based Trachoma Mapping in Six Evaluation Units of Papua New Guinea. Ophthalmic Epidemiol. 2016;23(sup1):22–31. pmid:27893297.

43. Sokana O, Macleod C, Jack K, Butcher R, Marks M, Willis R, et al. Mapping Trachoma in the Solomon Islands: Results of Three Baseline Population-Based Prevalence Surveys Conducted with the Global Trachoma Mapping Project. Ophthalmic Epidemiol. 2016;23(sup1):15–21. pmid:27937043.

44. Taleo F, Macleod CK, Marks M, Sokana O, Last A, Willis R, et al. Integrated Mapping of Yaws and Trachoma in the Five Northern-Most Provinces of Vanuatu. PLoS Negl Trop Dis. 2017;11(1):e0005267. pmid:28118354.

45. World Health Organization. Report of the 3rd global scientific meeting on trachoma. Johns Hopkins University, Baltimore, MA, 19–20 July 2010: Geneva: World Health Organization; 2010.

46. Global elimination of blinding trachoma. 51st World Health Assembly; May 16, 1998; Geneva: World Health Assembly; 1998.

47. World Health Organization. Validation of elimination of trachoma as a public health problem. WHO/HTM/NTD/2016.8. Geneva: World Health Organization, 2016.

48. Courtright P, Rotondo LA, MacArthur C, Jones I, Weaver A, Negash BK, et al. Strengthening the links between mapping, planning and global engagement for disease elimination: lessons learnt from trachoma. Br J Ophthalmol. 2018;102(10):1324–7. pmid:29907634; PubMed Central PMCID: PMC6173819.

49. Rajak SN, Habtamu E, Weiss HA, Bedri A, Gebre T, Bailey RL, et al. The clinical phenotype of trachomatous trichiasis in Ethiopia: not all trichiasis is due to entropion. Invest Ophthalmol Vis Sci. 2011;52(11):7974–80. pmid:21896855; PubMed Central PMCID: PMC3219424.

50. World Health Organization Alliance for the Global Elimination of Trachoma by 2020. Second Global Scientific Meeting on Trachomatous Trichiasis; Cape Town, 4–6 November 2015 (WHO/HTM/NTD/2016.5): Geneva: World Health Organization; 2016.

51. Ngondi J, Gebre T, Shargie EB, Graves PM, Ejigsemahu Y, Teferi T, et al. Risk factors for active trachoma in children and trichiasis in adults: a household survey in Amhara Regional State, Ethiopia. Trans R Soc Trop Med Hyg. 2008;102(5):432–8. pmid:18394663.

52. Burton MJ, Bowman RJ, Faal H, Aryee EA, Ikumapayi UN, Alexander ND, et al. The long-term natural history of trachomatous trichiasis in the Gambia. Invest Ophthalmol Vis Sci. 2006;47(3):847–52. pmid:16505016.

53. Jones BR. The prevention of blindness from trachoma. Trans Ophthalmol Soc U K. 1975;95(1):16–33. pmid:775692.

54. Schemann JF, Laffly D, Sacko D, Zephak G, Malvy D. Trichiasis and geoclimatic factors in Mali. Trans R Soc Trop Med Hyg. 2007;101(10):996–1003. pmid:17658570.

55. World Health Organization Strategic and Technical Advisory Group on Neglected Tropical Diseases. Design and validation of a trachomatous trichiasis-only survey (WHO/HTM/NTD/PCT/2017.08). Geneva: World Health Organization; 2018.